Glucocorticoids are routinely prescribed to treat a variety of diseases in human and veterinary medicine, including inflammatory, neoplastic, and immune-mediated disorders.1 Because glucocorticoids act on multiple organ systems and can cause adverse effects, concern exists regarding administration of these drugs to patients with comorbidities.
In particular, concern for use of glucocorticoids in patients with heart disease stems from reports of steroid-associated CHF in cats. In a retrospective study2 of cats with CHF secondary to hypertrophic cardiomyopathy, administration of glucocorticoids had recently preceded the diagnosis of CHF in 13 of 160 (8%) cats. In another study,3 12 cats with no known preexisting heart disease developed CHF a median of 4 days after administration of the long-acting injectable steroid methylprednisolone acetate (5 mg/kg, IM). A follow-up study4 by the same research group investigated potential mechanisms by which glucocorticoids might cause CHF. On the basis of findings of moderate to severe hyperglycemia and increased plasma volume following IM administration of methylprednisolone acetate to cats as monotherapy for dermatologic conditions, the authors concluded that an intravascular fluid shift resulting from glucocorticoid-induced hyperglycemia was the most likely mechanism for steroid-induced CHF in cats.4
The cardiovascular effects of orally administered anti-inflammatory doses of intermediate-acting glucocorticoids have recently been investigated in dogs and cats. In cats, oral administration of prednisolone (1 and 2 mg/kg, PO, q 24 h for 7 days each in a crossover-design study) caused no changes in blood glucose concentration, SAP, echocardiographic variables, or circulating concentrations of cardiac biomarkers.5 In dogs with allergic dermatitis, prednisone administration (1 mg/kg, PO, q 24 h) for 14 days caused no changes in blood glucose concentration or echocardiographic variables, but the treatment was associated with a clinically relevant increase in mean SAP after 7 days that represented a significantly greater change from baseline, compared with changes from baseline for a matched control group of healthy dogs.6 Similarly, high-dose glucocorticoid treatment in dogs (dexamethasone [0.5 mg/kg, PO, q 24 h] or hydrocortisone [8 mg/kg, PO, q 12 h]; both equivalent to 3 to 4 mg of prednisone/kg/d) has been reported to cause increased SAP,7,8 and systemic hypertension is common in people9 and dogs10 with hyperadrenocorticism. These results suggest that vasoconstriction and increased afterload, rather than changes in glucose homeostasis, may be clinically relevant effects of glucocorticoid treatment in dogs.
Despite the anecdotal and theoretical concern that glucocorticoid administration might potentiate CHF in cats, no data exist to suggest adverse effects of short-term anti-inflammatory administration of oral intermediate-acting glucocorticoids in patients with heart disease. In fact, it is possible that diuretic effects of glucocorticoids may be beneficial to patients with CHF. Results of several experimental studies11–20 in rats11,12,15,18,20 and sheep13,14 in vivo as well as in cultured bovine16 and rat17,19 cells in vitro have shown that glucocorticoid administration increases renal blood flow and GFR via both direct mechanisms (eg, dopamine, nitric oxide, and vasodilation of afferent arterioles)11–15 and indirect mechanisms (eg, potentiation of atrial natriuretic peptide).16–20 In clinical trials involving human patients with advanced decompensated CHF and diuretic resistance, prednisone administration potentiated diuresis and led to improved survival rate.21–24
Although increases in SAP have been reported following glucocorticoid treatment at various dosages and for different durations in dogs,6–8 it remains unclear whether the magnitude of change in SAP is related to dosages of these drugs. Additionally, glucocorticoids have been shown to increase GFR in people21,23 and rodents,11,12 but to the authors’ knowledge, a similar effect has not been found in dogs. The purpose of the study reported here was to determine whether a dose-response relationship exists for the effect of orally administered prednisone on hemodynamic and clinicopathologic variables in healthy dogs. This study was performed in the context of a larger investigation25 of ocular pharmacokinetics of prednisone after oral administration in the same healthy dogs. We hypothesized that short-term oral administration of prednisone would increase SAP and GFR in a dose-dependent manner in healthy dogs.
Materials and Methods
Animals
Eight systemically healthy purpose-bred Beagles were included in the study. All dogs were 1.5-year-old spayed females (body weight, 7 to 10 kg). Health prior to study enrollment was assessed through physical examination by a veterinarian and routine laboratory screening tests (CBC, serum biochemical analysis, and urinalysis). Absence of structural cardiac disease was confirmed prior to enrollment with transthoracic echocardiographya performed by a board-certified cardiologist (JLW) with standard 2–D, M-mode, color Doppler, and spectral Doppler echocardiography methods.26,27 An a priori sample size calculation indicated that 8 dogs would be needed to detect an increase from a baseline (pretreatment) SAP of 20 mm Hg after prednisone treatment, compared with results when dogs received no prednisone, with 90% power and an α of 0.05. These values were chosen on the basis of the reported change in SAP following 7 days of oral prednisone administration at a dose of 1 mg/kg, every 24 hours, in dogs with allergic dermatitis, compared with results for healthy control dogs that received no treatment.6 The study design was approved by the Iowa State University Institutional Animal Care and Use Committee (protocol No. 1-18-8692-K).
Study design
This study had a repeated-measures design, with each dog participating in 5 sequential experiments as follows: a control condition (no prednisone) and 4 prednisone treatments (0.5, 1, 2, and 4 mg/kg, PO, q 24 h). The treatments were given in ascending dose order without the use of random assignment. The dose of 4 mg/kg was chosen as the highest daily dose tested because it represents the high end of the dose range clinically indicated for canine immunosuppression1 and because prior studies7,8 have assessed selected clinicopathologic effects of equipotent doses of other glucocorticoids, including injectable dexamethasone.
The duration of each treatment was 5 consecutive days. Dogs were examined, and diagnostic tests were performed on days 1 (prior to prednisone administration) and 5 (30 minutes after prednisone administration). A treatment duration of 5 days was chosen on the basis of time needed to achieve a pharmacokinetic steady state for prednisone, and the timing of day 5 diagnostic tests was chosen to correspond to the time of maximal plasma concentration of prednisolone (bioactive product of prednisone) following oral administration of prednisone in dogs.28,29 A washout period of 9 days in which no prednisone was administered was provided between protocols. The 9-day washout period was chosen for convenience to allow a Monday through Friday schedule for treatment and data collection and represented an interval > 7 times the terminal elimination half-life of prednisone in dogs after oral administration (approx 1.4 hours)28,29 and was similar to the time required to reestablish normal hypothalamic-pituitary-adrenal function in people after short-term prednisone administration.30 In dogs, recovery of normal hypothalamic-pituitary-adrenal function has been demonstrated to occur ≤ 2 weeks after discontinuing prednisone at a dose of 0.55 mg/kg, PO, every 12 hours for 35 days.31
Dogs were fed their routine diet prior to examinations between 6:30 and 7:00 am, except for 1 dog that was offered food each morning but consistently consumed it later in the day, which was after completion of diagnostic testing on data collection days. Dogs were pair housed in the Laboratory Animal Resources unit at the Iowa State University College of Veterinary Medicine. Housing conditions were standardized with ambient temperatures of 18.9°C to 20°C, a 12-hour light cycle (6:00 am to 6:00 pm), and access to water ad libitum.
Data collection
On data collection days (days 1 and 5), the following tests were performed: measurement of body weight, heart rate, and SAP; a CBC and serum biochemical analysis; measurement of serum NT-proBNP concentration and urine specific gravity; and determination of fractional excretion of sodium and potassium, UPC, and GFR. Additional plasma from each collection day was stored for later analysis of plasma prednisolone concentration, free and total cortisol concentrations, and renin activity.
Dogs were brought from their living quarters into a dedicated procedure room for data collection. Dogs were allowed to acclimate to the procedure room, examination process, venipuncture, and SAP measurement protocols during 4 or 5 practice trials that took place 1 to 2 weeks prior to data collection. Dogs were weighed at each time point with the same digital scale. Systolic arterial blood pressure was measured in the quiet environment of the procedure room with gentle restraint. A standard noninvasive Doppler ultrasonic flow probe was used by the same 3 trained examiners (RLT, ALM, and YY) following standard methods.32 Consistent cuff sizes and patient positioning were used for each dog, and each SAP measurement was obtained from the right hind limb, except that 1 dog had the right forelimb used consistently for this purpose. A minimum of 3 consistent SAP measurements were obtained and mathematically averaged; up to 7 measurements were performed as needed to ensure stable measurements.
Venous blood samples were collected from an external jugular vein with 1-inch, 22-gauge needles attached to 12-mL chilled syringes. Approximately 8 mL of blood was collected, with 3 mL transferred into a 5-mL additive-free red-top tube and 2 mL transferred into a 3-mL EDTA-containing purple-top tube. Of the remaining blood, 2 mL was placed into a chilled 3-mL EDTA-containing tube prepared with 0.1 mL of aprotininb (bovine lung origination, 5 mg/mL) solution and stored on ice for < 30 minutes until use. Urine samples were collected by means of ultrasound-guided cystocentesis with 1-inch, 22-gauge needles attached to 3-mL syringes. Collected urine was transferred into 5-mL additive-free tubes and submitted immediately to the clinical pathological laboratory for testing. Chilled EDTA-treated samples with aprotinin additive were centrifuged at 4°C and 1,500 × g for 30 minutes. Plasma was transferred to cryovials and stored at, −80°C for later analysis of plasma renin activity, free and total plasma cortisol concentrations, and plasma prednisolone concentrations.
All CBCs, routine serum biochemical analyses, and urinalyses were performed at the Iowa State University Clinical Pathology Laboratory. Canine serum NT-proBNP measurements were performed at a commercial veterinary laboratory.c
An iohexol clearance technique was used to estimate GFR. The 0.75-inch, 21-gauge needle of a winged infusion set was inserted into a lateral saphenous or cephalic vein for iohexold (300 mg/kg) administration followed by a 2.5-mL heparinized saline (0.9% NaCl) solution flush. Blood samples (1.5 mL) were collected from the jugular vein 2, 3, and 4 hours after iohexol administration with a 1-inch, 22-gauge needle attached to a 3-mL syringe, and blood was placed into additive-free 5-mL tubes. Samples were centrifuged for 10 minutes at 500 × g, and serum was collected and stored at 3°C until analysis. Plasma iohexol concentration was measured by inductively coupled plasma mass spectrometry, and GFR calculations were performed as previously described33 at a university laboratory.e The interday and intraday coefficients of variation for iohexol analysis are provided (Supplementary Appendix S1, available at avmajournals.avma.org/doi/suppl/10.2460/ajvr.81.4.317).
The percentage change in plasma volume for each treatment was calculated by a method previously used to detect fluid balance alterations in cats following glucocorticoid treatment4,34 with the following equation:
where Hb represents hemoglobin concentration, D0 represents day 0, and D5 represents day 5.
The fractional excretion of sodium and potassium was calculated35 as follows:
where e represents the concentration of the electrolyte (in milliequivalents per liter) and c represents the concentration of creatinine (in milligrams per deciliter) in the specified sample type.
Batch analysis of plasma prednisolone concentrations was performed at the Iowa State University Veterinary Diagnostic Laboratory by means of liquid chromatography-mass spectrometry. Analysis of plasma free and bound cortisol concentrations was performed with equilibrium dialysis followed by liquid chromatography-tandem mass spectrometry in which cortisol-d4 was used as a surrogate analyte for calibration. Equilibrium dialysis was performed for 150 μL of each plasma sample dialyzed against 350 μL of PBS solution. Free and bound cortisol were separated by protein precipitation prior to liquid chromatography-tandem mass spectrometry. Total plasma cortisol concentration was calculated as the sum of free and bound cortisol concentrations. Plasma renin activity was determined by measuring plasma angiotensin I concentrations before and after incubation. A commercially available competitive binding peptide enzyme immunoassayf similar to other test kits previously validated for measurement of angiotensin I concentration in dogs36–38 was used.
Statistical analysis
Distribution of the data was assessed for normality by the Shapiro-Wilk test. All variables were found to be normally distributed; therefore, results were reported as mean ± SD. The percentage change from baseline (day 1 [prior to prednisone administration when applicable]) was computed for each variable after administration of the last treatment on day 5. Data for each response variable were analyzed with a linear mixed-effects modeling approach. Treatment, time, and the interaction between treatment and time were included in the model as fixed effects, and dog was included as a random effect. The significance of main effects and interactions was assessed with F tests, followed by diagnostic tests to check the model assumptions. If overall fixed effects were significant, post hoc pairwise comparisons were performed with a Tukey test to determine whether differences occurred between treatment conditions (treatment effect), pretreatment versus posttreatment time points (time effect), or the interaction between treatment and time. Additional linear mixed-effects models with post hoc multiple pairwise comparisons were performed to test for differences among baseline values for all experiments as an assessment for carryover effects of previous prednisone administration. Commercially available softwareg was used to perform all statistical analyses, and values of P < 0.05 were considered significant.
Results
The mean ± SD percentage change on day 5, compared with the baseline (day 1) value for the same treatment condition, was summarized for selected variables (Table 1). Changes in body weight, SAP, plasma volume, and serum NT-proBNP concentration did not differ significantly after administration of prednisone at any dose, compared with results for the same variables with the control condition (no prednisone). The highest mean SAP at any time point during the study was 143 mm Hg and occurred on day 1 (prior to prednisone administration) for the 1-mg/kg treatment condition. The percentage change from baseline in heart rate was significantly (P = 0.007) different when dogs received the 1-mg/kg dose of prednisone, compared with results for the control condition (−27% and 0.5%, respectively).
The GFR when dogs received prednisone at the 4-mg/kg dose was increased from baseline, compared with results for the control condition (P = 0.007; Table 1). A significant interaction between treatment condition and the change in UPC was observed, with a greater increase from baseline at the 4-mg/kg dose than for the control condition (P = 0.002) or doses of 0.5 mg/kg (P = 0.017) and 1 mg/kg (P = 0.035). Fractional excretion of sodium was decreased from baseline at doses of 0.5, 1, and 4 mg/kg (P = 0.003, P = 0.069, and P = 0.003, respectively), compared with results for the control condition, whereas changes in serum sodium concentration did not differ significantly among treatment conditions. Changes in urine specific gravity and fractional excretion of potassium also did not differ significantly among treatment conditions, but serum potassium concentration was increased from baseline when dogs received the 1-mg/kg dose of prednisone, compared with results for the control condition (P = 0.023).
Mean ± SD day 5 percentage change from baseline (day 1) values for target variables in 8 healthy Beagles that underwent five 5-day experiments (no prednisone treatment [control condition] and prednisone administration at 0.5, 1, 2, and 4 mg/kg, PO, q 24 h) with a 9-day washout period between protocols.
| Change from baseline (%) | |||||
|---|---|---|---|---|---|
| Variable | Control | 0.5 mg/kg | 1 mg/kg | 2 mg/kg | 4 mg/kg |
| SAP (mm Hg) | −2.9 ± 16.7 | 4.8 ± 33.3 | 1.3 ± 19.0 | 3.2 ± 8.0 | 8.8 ± 11.7 |
| GFR (mL/min/kg) | −14.6 ± 21.4 | 20.5 ± 60.2 | 20.9 ± 49.3 | 23.1 ± 3048 | 50.28 ± 70.3* |
| UPC | −23.4 ± 24.8 | −334 ± 20.3 | −20.8 ± 37.6 | 204 ± 78.7 | 33.6 ± 30.6*†‡ |
| Fractional excretion of sodium | −10.9 ± 16.6 | −634 ± 40.3* | –45.5 ± 34.5* | −21.0 ± 104.5† | −57.4 ± 91.0*§ |
| Serum analytes | |||||
| Sodium (mEq/L) | −0.2 ± 1.1 | −04 ± 1.5 | −0.5 ± 1.3 | −0.3 ± 1.5 | −1.9 ± 2.9 |
| Potassium (mEq/L) | 3.3 ± 4.2 | −0.2 ± 9.2 | 4.8 ± 5.1* | −2.6 ± 8.0 | −0.7 ± 13.2 |
| Chloride (mEq/L) | −0.9 ± 1.5 | −7.0 ± 2.8* | −6.1 ± 2.9* | −7.9 ± 3.3* | −8.0 ± 0.9* |
| Magnesium (mg/dL) | 3.0 ± 2.4 | 16.3 ± 6.3* | 13.3 ± 9.2* | 20.3 ± 13.3* | 10.2 ± 10.1*§ |
| Glucose (mg/dL) | −3.4 ± 10.8 | −8.7 ± 17.0 | −15.4 ± 22.8 | −1.8 ± 6.3 | 17.5 ± 19.6*†‡§ |
| Total protein (g/dL) | 4.3 ± 3.9 | 11.2 ± 5.5* | 8.5 ± 3.3* | 12.5 ± 4.2* | 11.2 ± 3.7* |
| Albumin (g/dL) | 4.9 ± 4.0 | 17.0 ± 7.5* | 14.8 ± 4.6* | 17.5 ± 6.0* | 15.9 ± 5.1* |
| Alkaline phosphatase (U/L) | −1.2 ± 10.7 | 71.5 ± 27.9* | 66.2 ± 25.0* | 106.6 ± 69.4* | 112.0 ± 54.9* |
| Cholesterol (mg/dL) | 3.0 ± 5.6 | 4.2 ± 6.8 | 9.3 ± 6.5 | 14.9 ± 11.6* | 6.8 ± 8.7 |
| Triglycerides (mg/dL) | −16.6 ± 30.0 | 82.7 ± 165.0* | 185.7 ± 182.9* | 127.5 ± 182.7* | 145.0 ± 146.6* |
| Plasma total cortisol (ng/mL) | 67.9 ± 122.7 | −714 ± 41.5* | −92.9 ± 18.9* | −100 ± 0* | −100 ± 0* |
Units provided are for the variables as measured or calculated. Within a row, symbols indicate significant (P < 0.05) differences in the change from baseline between treatments (linear mixed-effects model with post hoc pairwise comparisons [Tukey test]).
Result is significantly different, compared with that for the control condition.
Result is significantly different from the result for the 0.5-mg/kg dose.
Result is significantly different from that for the 1-mg/kg dose.
Result is significantly different from that for the 2-mg/kg dose.
Serum glucose concentration was increased from baseline when dogs received the 4-mg/kg dose of prednisone, compared with the changes observed for the control condition (P = 0.023) and for the doses of 0.5 mg/kg (P = 0.002), 1 mg/kg (P = 0.029), and 2 mg/kg (P = 0.019; Table 1). All doses of prednisone were associated with decreases in serum chloride concentration and increases in serum albumin, total protein, triglyceride, and magnesium concentrations and alkaline phosphatase activities, compared with results for the control condition (P < 0.05 for all comparisons). Changes in alanine aminotransferase activities did not differ significantly among treatment conditions, and changes in serum bicarbonate and cholesterol concentrations each differed significantly from results for the control condition at only 1 prednisone dose, with increases from baseline in bicarbonate concentration (17.6% vs 2.9%; P = 0.009) when dogs received the 0.5-mg/kg dose and in cholesterol concentration (P = 0.036) when they were administered the 2-mg/kg dose. Other than triglyceride concentrations, all changes in routine biochemical analytes associated with prednisone treatment were mild, with mean values for all dogs remaining well within the respective reference ranges.
The CBC results revealed that platelet count was significantly (P = 0.036) increased from baseline (12.9%) and band neutrophil count was significantly (P < 0.01) decreased from baseline (−94%) at a prednisone dose of 0.5 mg/kg, compared with changes detected for the control condition (−2.5 and 0%, respectively). Mean changes in these variables were mild, and mean values remained within the respective reference intervals. No other changes in hematologic variables differed significantly from those for the control condition.
Changes from baseline in plasma total cortisol concentration differed from changes for the control condition at all doses of prednisone (P < 0.01); decreases were observed when dogs received prednisone treatment at any dose, whereas a small increase from baseline was observed in untreated dogs. No significant changes in plasma free cortisol concentration or plasma angiotensin I concentration (measure of renin activity) were associated with prednisone treatment. No prednisolone was detectable in plasma at baseline for any treatment condition or in day 5 samples for the control condition. Mean plasma prednisolone concentrations on day 5 were 20 ± 25 ng/mL, 45 ± 85 ng/mL, 121 ± 213 ng/mL, and 137 ± 163 ng/mL when dogs received 0.5, 1, 2, and 4 mg/kg of prednisone, respectively. Only the change from baseline in plasma prednisolone concentration at the 4-mg/kg dose (156%) was different from the result for the control condition (0%; P = 0.032).
Baseline values for 4 variables differed significantly among the 4 prednisone treatments. On pairwise comparisons, the baseline circulating band neutrophil concentration was higher at the start of the 0.5-mg/kg treatment period (0.19 × 103 cells/μL) than for all subsequent treatment periods (0.01 × 103 cells/μL, 0.02 × 103 cells/μL, and 0.0 × 103 cells/μL for the 1-, 2-, and 4-mg/kg treatments, respectively; P < 0.001). Baseline serum cholesterol concentration for the 2-mg/kg treatment (129 mg/dL) was lower than the baseline values for the control condition (155 mg/dL; P < 0.001) and the 0.5-mg/kg (151 mg/dL; P < 0.001) and 1-mg/kg (143 mg/dL; P = 0.023) treatments. Serum total protein (6.1 g/dL) and albumin (3.4 g/dL) concentrations at baseline were higher for 4-mg/kg treatment, compared with those for the 0.5-mg/kg treatment only (total protein, 5.8 g/dL [P = 0.036]; albumin, 3.2 g/dL [P = 0.039]). Pairwise comparisons revealed no significant differences between baseline values in the 4 prednisone treatments for any of the target variables in the study, including serum glucose concentration, SAP, and GFR. Linear mixed model analysis identified no significant interactions between time and treatment for any variables measured, including these target variables.
Discussion
Results of the present study revealed only mild clinicopathologic and hemodynamic changes in healthy dogs following short-term (5-day) oral administration of prednisone at each of 4 doses (0.5, 1, 2, and 4 mg/kg) every 24 hours, with a 9-day washout period between protocols. The most clinically relevant changes occurred with the highest dose administered.
Contrary to the first aspect of our hypothesis, changes in mean SAP from baseline measurements (day 1, prior to drug administration for the prednisone treatment protocols) did not differ significantly from those for the control condition at any of the prednisone doses administered. This result was in contrast to findings of a previous study6 in which the mean change from baseline SAP for dogs that received prednisone (1 mg/kg, PO, q 24 h) for treatment of dermatologic conditions was significantly greater than that for a matched group of untreated healthy control dogs; furthermore, mean SAP values in that study6 eclipsed 160 mm Hg, crossing the threshold for the definition of systemic hypertension in dogs. In contrast, mean SAP measurements in the present study did not exceed 160 mm Hg at any prednisone dose. Results of other investigations have revealed significant increases in SAP following short-term7 or long-term8 administration of high doses of glucocorticoids (equivalent to 3 to 4 mg of prednisone/kg, q 24 h), but SAP measurements in those studies remained < 150 mm Hg. These disparate SAP results among studies may be related to differences in standardization of living conditions and acclimation to experimental protocols, which can affect relative sympathetic nervous system activation during SAP measurement (situational hypertension,39 sometimes termed white-coat hypertension or white-coat effect). On the basis of information from serial Doppler measurements in laboratory-housed Beagles that shows SAP progressively decreases over the first 4 measurements and then plateaus,40 the present investigation provided 4 or 5 acclimation sessions prior to data collection. Such an acclimation period was not feasible in the previous study6 involving client-owned dogs. Marked increases in SAP can lead to clinically relevant increases in left ventricular afterload and secondary structural and functional cardiac abnormalities, specifically concentric left ventricular hypertrophy, dilation of the ascending aorta, coronary artery ischemia, and left ventricular diastolic dysfunction, as has been described for cats.41,42 Results of the present study suggested less concern than previously reported regarding the potential for short-term glucocorticoid therapy to augment afterload and thus exacerbate preexisting heart disease in dogs.
In support of the second aspect of our hypothesis, a substantial increase from baseline in mean GFR was detected after 5 days of prednisone administration at 4 mg/kg, and this change was significantly different from the change in GFR (a mean decrease) observed for the control condition. These data confirmed prior evidence from other species that glucocorticoid administration increases renal blood flow and GFR through numerous direct (eg, vasodilation of the afferent arteriole mediated by dopamine and nitric oxide) and indirect (eg, potentiation of atrial natriuretic peptide) mechanisms.11–20 This finding corroborated the notion that oral prednisone administration may potentiate diuresis and thus may have a potential therapeutic role in patients with advanced CHF and diuretic resistance. However, the described increase in GFR for dogs of the present study was significantly different from changes for the control condition only at the highest (immunosuppressive) prednisone dose, which is not likely to be feasible for long-term administration to patients with CHF. However, it is important to note that the diuretic effects of prednisone are not limited to changes in GFR; glucocorticoids also inhibit release of antidiuretic hormone from the posterior pituitary and interfere with receptor binding and signal transduction of antidiuretic hormone in the collecting duct.43 Although effects on antidiuretic hormone were not specifically assessed in this study, it is interesting to note that changes in urine specific gravity with administration of prednisone at any dose did not differ significantly from the changes observed for the control condition. Analysis of changes in UPC yielded findings complementary to GFR results. A previous report8 described an increase in urinary protein excretion with administration of prednisone, a finding attributable to increased GFR causing more blood to be filtered at the glomeruli and therefore more protein to be excreted. Interestingly, mean increases in UPC were observed in the present study only at the 2- and 4-mg/kg doses of prednisone, and the change from baseline differed significantly only for the 4-mg/kg dose, compared with results for the control condition and the 0.5- and 1-mg/kg doses.
Mineralocorticoid effects from prednisone are expected to cause an increase in tubular resorption of sodium and increased tubular secretion of potassium.44 This mechanism was partially supported by results of the present study, in which mean fractional excretion of sodium at prednisone doses of 0.5, 1, and 4 mg/kg was significantly decreased from baseline, compared with changes observed for the control condition. The decrease in sodium excretion was particularly interesting and unexpected in the context of increased GFR because higher flow rates in the proximal tubule typically interfere with passive concentration-dependent absorption from the proximal tubule and lead to increased sodium loss.45 Although the mechanism for the dual findings of increased GFR and decreased fractional excretion of sodium in our study was unknown, potential contributing factors might include glomerulotubular balance (increased tubular flow causing increased sodium absorption in the distal nephron)46 or a direct mineralocorticoid effect of prednisone. Such counterbalance between increased GFR and decreased fractional sodium excretion might also explain the lack of change in plasma volume following prednisone administration. The changes from baseline in mean serum sodium and potassium measurements did not clearly reflect the observed patterns in urinary excretion of these electrolytes; mean sodium concentration changes did not differ significantly from those for the control condition at any prednisone dose, and mean potassium concentration was increased from baseline when dogs received 1 mg of prednisone/kg when compared with results for the control condition. These results may have been attributable to homeostatic mechanisms to stabilize plasma electrolyte concentrations in the context of variable urinary excretion. Overall, results of the study reported here supported previous reports4–6 that glucocorticoids may exert mild mineralocorticoid effects at the level of the kidney without consistent alterations in serum electrolytes.
Mean serum glucose concentrations increased from baseline, compared with findings for the control condition, at the highest prednisone dose (4 mg/kg) only. Although the increase in glucose concentrations was mild and values did not exceed the upper reference limit, this result was similar to previous findings in cats administered immunosuppressive doses of prednisolone (2 to 4.4 mg/kg, PO, q 24 h) for 1 to 8 weeks.47,48 Reports49–53 of hyperglycemia or glucose intolerance following glucocorticoid administration in dogs have been inconsistent, and an oral dose of 4 mg/kg, every 24 hours, has not specifically been investigated to the authors’ knowledge. Results of a previous study6 from our research group reveal no changes in serum glucose concentrations with short-term prednisone treatment at an anti-inflammatory dose (1 mg/kg) in dogs. Results of the present study suggested that steroid-induced increases in circulating glucose concentrations may be a potential adverse effect of high-dose (immunosuppressive) prednisone treatment in dogs. However, the clinical relevance of this finding was unclear because serum glucose concentrations remained within the reference range and were not associated with a concomitant increase in plasma volume. Overall, these data suggested a minimal risk of adverse cardiovascular outcomes (increased intravascular volume and potentially precipitation of CHF), in contrast to findings described for cats that received injectable methylprednisolone acetate.4
Several expected glucocorticoid-associated changes in CBC and serum biochemical data were noted at various doses of prednisone administration in the present study. Mean chloride concentrations decreased from baseline with all prednisone treatments as expected, with significant differences from the control condition at all doses and subjectively greater decreases at the 2 highest doses. This finding was considered attributable to glucocorticoid stimulation of endogenous organic acid production and renal tubular secretion of hydrogen and accompanying chloride ions.54,55 Enzymatic and lipid-related changes associated with administration of prednisone, compared with changes for the control condition, included the anticipated increases in mean alkaline phosphatase activity and triglyceride concentrations. Mean albumin and total protein concentrations were consistently increased from baseline, compared with results for the control condition, at all doses of prednisone; these findings may have been attributable to volume contraction (although no significant differences in changes from baseline for plasma volume were associated with treatment in the present study, compared with control condition results) or increased hepatic synthesis of albumin induced by glucocorticoids.31,56–58 Plasma mean total cortisol concentration decreased from baseline at all doses of prednisone when compared with results for the control condition, an anticipated result owing to suppression of the hypothalamic-pituitary-adrenal axis.31 These biochemical changes were consistent with steroid administration1,44,49,59,60 and are not considered relevant factors in the development or progression of heart disease in dogs.
Changes in mean plasma renin activity did not differ from findings for the control condition at any prednisone dose, suggesting no obvious glucocorticoid-associated activation of the RAAS in healthy dogs of the present study. These findings were in agreement with results of previous investigations of RAAS activation in dogs with exogenous (dexamethasone-treated) or endogenous (hyperadrenocorticism) glucocorticoid excess. One such study7 found no change from baseline in plasma renin activity of healthy dogs after 10 days of treatment with dexamethasone (equivalent to 3.5 mg of prednisone/kg, q 24 h). Results of another study61 indicated no difference in plasma aldosterone concentrations or plasma renin activity between healthy dogs and dogs with hyperadrenocorticism. Activation of the RAAS has been suggested as a potential mechanism for glucocorticoid-associated hypertension in rodents and people9,62; however, no evidence of increased SAP or RAAS activation was identified at any prednisone dose in this study of healthy dogs. It is possible that sampling for plasma renin activity used in the present study was performed too infrequently to detect transient or oscillating changes in RAAS activation. Alternatively, competing pathophysiologic effects of high-dose glucocorticoid administration (increased GFR with a concurrent increase in renal sodium retention) may result in no net effect on the RAAS.
The present study had several limitations. First, the number of dogs was small and the study was powered to detect differences in SAP specifically; therefore, it may have been underpowered to detect differences in other outcome variables expected to change with glucocorticoid treatment, such as urine specific gravity. In addition, this study was part of a larger investigation of ocular pharmacokinetics of prednisone in dogs, which dictated certain aspects of the study design. The 5-day duration of prednisone administration was selected on the basis of known pharmacokinetic attributes (eg, peak circulating concentrations and terminal elimination half-life) of prednisolone; however, our sampling protocol may have failed to capture a cumulative 5-dose effect and may have missed genomic (gene expression) effects of prednisone.63 Clinicopathologic or hemodynamic changes may have been different, or greater changes may have developed, with long-term glucocorticoid administration. The data from the present report cannot necessarily be extrapolated to long-term glucocorticoid administration, which is common for dogs with chronic inflammatory or immune-mediated diseases. Another limitation was that all dogs included in this study were free of evidence of cardiac disease, similar to animals enrolled in previous investigations of the cardiovascular effects of glucocorticoids in dogs and cats with dermatologic disease.5,6 Further studies are needed to investigate whether cardiovascular effects of varying doses of prednisone treatment might differ in dogs with underlying cardiac disease and whether those changes could cause clinically relevant disease progression. Investigators of the present study were not blinded to the treatment condition or prednisone dose when performing data collection, which could have introduced bias in heart rate or SAP measurements. Further, although living conditions of the study dogs were highly standardized and the dogs had access to water ad libitum, food intake, water consumption, and urine production were not quantified; thus, inconsistent food or water consumption might have influenced the results for GFR or electrolyte excretion in individual animals. We also could not confirm with the manufacturer whether the commercially available angiotensin I test kit used to measure plasma renin activity in the study has been specifically validated for use with canine samples. An additional limitation was that plasma samples were stored at −80°C for periods of 2 to 6 months prior to measurement of certain analytes, and we could not exclude the possibility of altered stability or sample degradation.
Perhaps the most important limitation of the present study was that the 4 prednisone doses were assessed in a predetermined order, beginning with the lowest and ending with the highest daily dose and having only 9-day washout periods between experiments. The protocol was based on the larger pharmacokinetic study design, and the 9-day washout period was chosen to allow clearance of plasma prednisolone. However, longer-term effects of glucocorticoids could have been persistent from prior treatments, and this washout period may have been insufficient to allow recovery of the hypothalamic-pituitary-adrenal axis. Specifically, from a statistical perspective, the relatively short washout period and lack of randomization of treatment order could have introduced the confounding variable of time (the chronological order of treatments) as a potential cause of changes in clinicopathologic variables. Our statistical models included time as a fixed effect, and post hoc pairwise testing did not identify significant differences between baseline values for target variables across treatment conditions, suggesting a minimal effect of time or carryover effect of previous prednisone administration. However, because of the small sample size and potential lack of statistical power, the fact that our statistical models did not identify an effect of time (or interaction between time and treatment) did not exclude the possibility of such an effect.
The present study demonstrated only a limited dose-response effect of short-term glucocorticoid administration on target variables in healthy dogs; only at the highest prednisone dose studied (4 mg/kg, q 24 h) did the increase from baseline in serum glucose, GFR, and UPC differ from that of other treatment conditions. Further evaluation of longer-term glucocorticoid administration in healthy dogs, as well as glucocorticoid treatment in dogs with preexisting cardiac disease, is warranted to investigate potential adverse effects and potential use for adjunctive diuresis.
Acknowledgments
Supported by a VCS Incentive Grant and an ACVIM Resident Research Award. Financial supporters had no involvement in study design, data analysis and interpretation, or writing and publication of the manuscript.
The authors declare that there were no conflicts of interest. The authors thank Lori Moran for technical assistance.
ABBREVIATIONS
| CHF | Congestive heart failure |
| GFR | Glomerular filtration rate |
| NT-proBNP | N-terminal pro B–type natriuretic peptide |
| RAAS | Renin-angiotensin-aldosterone system |
| SAP | Systolic arterial blood pressure |
| UPC | Urine protein-to-creatinine ratio |
Footnotes
CX50 Philips Ultrasound with FUS8392 S8-3 transducer, Philips Healthcare, Andover, Mass.
Aprotinin lyophilized powder, 3 to 8 TIU/mg, Sigma-Aldrich Corp, St Louis, Mo.
Idexx Laboratories, Westbrook, Me.
Omnipaque, 350 mg/mL, GE Healthcare, Chicago, Ill.
Michigan State University Diagnostic Center for Population and Animal Health, Lansing, Mich.
Angiotensin I EIA Kit, catalog No. S-1188.0001, Peninsula Laboratories International, San Carlos, Calif.
R software, version 3.5.2, R Foundation for Statistical Computing, Vienna, Austria.
References
1. Plumb DC. Plumb's veterinary drug handbook. 8th ed. Hoboken, NJ: Wiley-Blackwell, 2015;1202–1208.
2. Rush JE, Freeman LM, Fenollosa NK, et al. Population and survival characteristics of cats with hypertrophic cardiomyopathy: 260 cases (1990–1999). J Am Vet Med Assoc 2002;220:202–207.
3. Smith SA, Tobias AH, Fine DM, et al. Corticosteroid-associated congestive heart failure in 12 cats. Int J Appl Res Vet Med 2004;2:159–170.
4. Ployngam T, Tobias AH, Smith SA, et al. Hemodynamic effects of methylprednisolone acetate administration in cats. Am J Vet Res 2006;67:583–587.
5. Khelik IA, Berger DJ, Mochel JP, et al. Clinicopathologic, hemodynamic, and echocardiographic effects of short-term oral administration of anti-inflammatory doses of prednisolone to systemically normal cats. Am J Vet Res 2019;80:743–755.
6. Masters AK, Berger DJ, Ware WA, et al. Effects of short-term anti-inflammatory glucocorticoid treatment on clinicopathologic, echocardiographic, and hemodynamic variables in systemically healthy dogs. Am J Vet Res 2018;79:411–423.
7. Nakamoto H, Suzuki H, Kageyama Y, et al. Characterization of alterations of hemodynamics and neuroendocrine hormones in dexamethasone induced hypertension in dogs. Clin Exp Hypertens A 1991;13:587–606.
8. Schellenberg S, Mettler M, Gentilini F, et al. The effects of hydrocortisone on systemic arterial blood pressure and urinary protein excretion in dogs. J Vet Intern Med 2008;22:273–281.
9. Saruta T, Suzuki H, Handa M, et al. Multiple factors contribute to the pathogenesis of hypertension in Cushing's syndrome. J Clin Endocrinol Metab 1986;62:275–279.
10. Reusch CE, Schellenberg S, Wenger M. Endocrine hypertension in small animals. Vet Clin North Am Small Anim Pract 2010;40:335–352.
11. Baylis C, Brenner BM. Mechanism of the glucocorticoid-induced increase in glomerular filtration rate. Am J Physiol 1978;234:F166–F170.
12. Baylis C, Handa RK, Sorkin M. Glucocorticoids and control of glomerular filtration rate. Semin Nephrol 1990;10:320–329.
13. De Matteo R, May CN. Glucocorticoid-induced renal vasodilatation is mediated by a direct renal action involving nitric oxide. Am J Physiol 1997;273:R1972–R1979.
14. de Matteo R, May CN. Inhibition of prostaglandin and nitric oxide synthesis prevents cortisol-induced renal vasodilatation in sheep. Am J Physiol 1999;276:R1125–R1131.
15. Aguirre JA, Ibarra FR, Barontini M, et al. Effect of glucocorticoids on renal dopamine production. Eur J Pharmacol 1999;370:271–278.
16. Lanier-Smith KL, Currie MG. Effect of glucocorticoids on the binding of atrial natriuretic peptide to endothelial cells. Eur J Pharmacol 1990;178:105–109.
17. Kanda K, Ogawa K, Miyamoto N, et al. Potentiation of atrial natriuretic peptide-stimulated cyclic guanosine monophosphate formation by glucocorticoids in cultured rat renal cells. Br J Pharmacol 1989;96:795–800.
18. Garcia R, Debinski W, Gutkowska J, et al. Gluco- and mineralocorticoids may regulate the natriuretic effect and the synthesis and release of atrial natriuretic factor by the rat atria in vivo. Biochem Biophys Res Commun 1985;131:806–814.
19. Liu C, Chen Y, Kang Y, et al. Glucocorticoids improve renal responsiveness to atrial natriuretic peptide by up-regulating natriuretic peptide receptor-A expression in the renal inner medullary collecting duct in decompensated heart failure. J Pharmacol Exp Ther 2011;339:203–209.
20. Liu C, Guan J, Kang Y, et al. Inhibition of dehydration-induced water intake by glucocorticoids is associated with activation of hypothalamic natriuretic peptide receptor-A in rat. PLoS One 2010;5:e15607.
21. Liu C, Liu G, Zhou C, et al. Potent diuretic effects of prednisone in heart failure patients with refractory diuretic resistance. Can J Cardiol 2007;23:865–868.
22. Liu C, Liu K, COPE-ADHF Study Group. Cardiac outcome prevention effectiveness of glucocorticoids in acute decompensated heart failure: COPE-ADHF study. J Cardiovasc Pharmacol 2014;63:333–338.
23. Zhang H, Liu C, Ji Z, et al. Prednisone adding to usual care treatment for refractory decompensated congestive heart failure. Int Heart J 2008;49:587–595.
24. Liu C, Chen H, Zhou C, et al. Potent potentiating diuretic effects of prednisone in congestive heart failure. J Cardiovasc Pharmacol 2006;48:173–176.
25. Sebbag L, Yan Y, Smith JS, et al. Tear fluid pharmacokinetics following oral prednisone administration in dogs with and without conjunctivitis. J Ocul Pharmacol Ther 2019;35:341–349.
26. Thomas WP. Two-dimensional, real-time echocardiography in the dog. Vet Radiol 1984;25:50–64.
27. Thomas WP, Gaber CE, Jacobs GJ, et al. Recommendations for standards in transthoracic two-dimensional echocardiography in the dog and cat. Echocardiography Committee of the Specialty of Cardiology, American College of Veterinary Internal Medicine. J Vet Intern Med 1993;7:247–252.
28. El Dareer SM, Struck RF, White VM, et al. Distribution and metabolism of prednisone in mice, dogs, and monkeys. Cancer Treat Rep 1977;61:1279–1289.
29. Colburn WA, Sibley CR, Buller RH. Comparative serum prednisone and prednisolone concentrations following prednisone or prednisolone administration to Beagle dogs. J Pharm Sci 1976;65:997–1001.
30. Carella MJ, Srivastava LS, Gossain VV, et al. Hypothalamic-pituitary-adrenal function one week after a short burst of steroid therapy. J Clin Endocrinol Metab 1993;76:1188–1191.
31. Moore GE, Hoenig M. Duration of pituitary and adrenocortical suppression after long-term administration of anti-inflammatory doses of prednisone in dogs. Am J Vet Res 1992;53:716–720.
32. Henik RA, Dolson MK, Wenholz LJ. How to obtain a blood pressure measurement. Clin Tech Small Anim Pract 2005;20:144–150.
33. Innis CJ, Kennedy A, McGowan JP, et al. Glomerular filtration rates of naturally cold-stunned Kemp's Ridley turtles (Lepidochelys kempii): comparison of initial vs. convalescent values. J Herpetological Med Surg 2017;26:100–103.
34. Harrison MH. Effects on thermal stress and exercise on blood volume in humans. Physiol Rev 1985;65:149–209.
35. Steiner RW. Interpreting the fractional excretion of sodium. Am J Med 1984;77:699–702.
36. Kohara K, Brosnihan KB, Chappell MC, et al. Angiotensin-(1–7): a member of circulating angiotensin peptides. Hypertension 1991;17:131–138.
37. Aikawa T, Suzuki S, Murayama M, et al. Enzyme immunoassay of angiotensin I. Endocrinology 1979;105:1–6.
38. Brosnihan KB, Chappell MC. Measurement of angiotensin peptides: HPLC-RIA. Methods Mol Biol 2017;1527:81–99.
39. Acierno MJ, Brown S, Coleman AE, et al. ACVIM consensus statement: guidelines for the identification, evaluation, and management of systemic hypertension in dogs and cats. J Vet Intern Med 2018;32:1803–1822.
40. Schellenberg S, Glaus TM, Reusch CE. Effect of long-term adaptation on indirect measurements of systolic blood pressure in conscious untrained Beagles. Vet Rec 2007;161:418–421.
41. Nelson L, Reidesel E, Ware WA, et al. Echocardiographic and radiographic changes associated with systemic hypertension in cats. J Vet Intern Med 2002;16:418–425.
42. Henik RA, Stepien RL, Bortnowski HB. Spectrum of M-mode echocardiographic abnormalities in 75 cats with systemic hypertension. J Am Anim Hosp Assoc 2004;40:359–363.
43. Bähr V, Franzen N, Oelkers W, et al. Effect of exogenous glucocorticoid on osmotically stimulated antidiuretic hormone secretion and on water reabsorption in man. Eur J Endocrinol 2006;155:845–848.
44. Stockham SL, Scott MA. Fundamentals of veterinary clinical pathology. 2nd ed. Ames, Iowa: Wiley-Blackwell, 2008;86, 390, 770–77.
45. Buckalew VM Jr, Walker BR, Puschett JB, et al. Effects of increased sodium delivery on distal tubular sodium reabsorption with and without volume expansion in man. J Clin Invest 1970;49:2336–2344.
46. Morimoto T, Liu W, Woda C, et al. Mechanism underlying flow stimulation of sodium absorption in the mammalian collecting duct. Am J Physiol Renal Physiol 2006;291:F663–F669.
47. Lowe AD, Graves TK, Campbell KL, et al. A pilot study comparing the diabetogenic effects of dexamethasone and prednisolone in cats. J Am Anim Hosp Assoc 2009;45:215–224.
48. Middleton DJ, Watson AD. Glucose intolerance in cats given short-term therapies of prednisolone and megestrol acetate. Am J Vet Res 1985;46:2623–2625.
49. Wolfsheimer KJ, Flory W, Williams MD. Effects of prednisolone on glucose tolerance and insulin secretion in the dog. Am J Vet Res 1986;47:1011–1014.
50. Moore GE, Hoenig M. Effects of orally administered prednisone on glucose tolerance and insulin secretion in clinically normal dogs. Am J Vet Res 1993;54:126–129.
51. Kovalik M, Thoday KL, Evans H, et al. Prednisolone is associated with an increase in serum insulin but not serum fructosamine concentrations in dogs with atopic dermatitis. Vet J 2012;192:212–216.
52. Greve T, Dayton AD, Anderson NV. Acute pancreatitis with coexistent diabetes mellitus: an experimental study in the dog. Am J Vet Res 1973;34:939–945.
53. Peterson ME, Altszuler N, Nichols CE. Decreased insulin sensitivity and glucose tolerance in spontaneous canine hyperadrenocorticism. Res Vet Sci 1984;36:177–182.
54. Hulter HN, Licht JH, Bonner EL Jr, et al. Effects of glucocorticoid steroids on renal and systemic acid-base metabolism. Am J Physiol 1980;239:F30–F43.
55. Hulter HN, Sigala JF, Sebastian A. Effects of dexamethasone on renal and systemic acid-base metabolism. Kidney Int 1981;20:43–49.
56. Rothschild MA, Oratz M, Schreiber SS. Albumin synthesis. N Engl J Med 1972;286:748–757.
57. Rothschild MA, Schreiber SS, Oratz M, et al. The effects of adrenocortical hormones on albumin metabolism studied with albumin-I 131. J Clin Invest 1958;37:1229–1235.
58. Rothschild MA, Oratz M, Wimer E, et al. Studies on albumin synthesis: the effects of dextran and cortisone on albumin metabolism in rabbits studied with albumin-I-131. J Clin Invest 1961;40:545–554.
59. Moore GE, Mahaffey EA, Hoenig M. Hematologic and serum biochemical effects of long-term administration of anti-inflammatory doses of prednisone in dogs. Am J Vet Res 1992;53:1033–1037.
60. Angeli A, Bisbocci D, Melo F, et al. Relative competition of corticosterone, cortisol, cortisone, 11-dexycortisol and prednisolone with (1,2,3H)-cortisol in various protein-binding radioassay systems. Clin Chim Acta 1975;61:279–286.
61. Javadi S, Kooistra HS, Mol JA, et al. Plasma aldosterone concentrations and plasma renin activity in healthy dogs and dogs with hyperadrenocorticism. Vet Rec 2003;153:521–525.
62. Suzuki H, Handa M, Kondo K, et al. Role of renin-angiotensin system in glucocorticoid hypertension in rats. Am J Physiol 1982;243:E48–E51.
63. Stahn C, Buttgereit F. Genomic and nongenomic effects of glucocorticoids. Nat Clin Pract Rheumatol 2008;4:525–533.
